DYNAMIC SERVICE MARKETPLACE QUERY BUILDING AND DECOMPOSITION SYSTEM

    公开(公告)号:US20240143599A1

    公开(公告)日:2024-05-02

    申请号:US18051180

    申请日:2022-10-31

    CPC classification number: G06F16/24575 G01C21/3841 G06F16/24539 G06Q30/0609

    Abstract: A resource sharing marketplace platform includes a back-office server in wireless communication with a controller adapted to run a plurality of integrated systems, and is adapted to authenticate identification of the user and facilitate user log-in, receive a request from a user, establish dynamic parameters related to the request, break down the request from the user into a plurality of serviceable micro-queries by decomposing at least one request for multivariate data into a plurality of requests for multivariate and univariate data, and develop a contract including key clauses and reward constraints, identify, from a group of registered service providers, a sub-group of registered service providers that are capable of providing data related to at least one micro-query, and send a request for data to the sub-group, collect data from a plurality of acceptable service providers, and repack the individual data to provide a response to the user.

    System and method of selecting a computational platform

    公开(公告)号:US11023351B2

    公开(公告)日:2021-06-01

    申请号:US15445435

    申请日:2017-02-28

    Abstract: Methods and apparatus are provided for selecting a computational platform. The method includes collecting platform feature data for each of a plurality of existing computational platforms in a particular class of computational platforms. The platform feature data for each of the plurality of existing computational platforms includes a plurality of design parameters and a performance measurement for the computational platform to complete a benchmarking software component. The method further includes training a mathematical model using the collected platform feature data to devise a predictive model for predicting a predicted performance measurement for a computational platform in the particular class of computational platforms when provided with design parameters, predicting, with the predictive model, the predicted performance measurement of a new computational platform that is not available for testing, and selecting a computational platform using the predicted performance measurement.

Patent Agency Ranking